Content
61%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is highly actionable with real, executable examples and a clear routing decision tree, supported by well-signaled one-level-deep reference files. Its weaknesses are generic best-practice padding, missing validation checkpoints for destructive/batch operations, and detailed code that could live in references.
Suggestions
Replace the generic Best Practices and Troubleshooting lists with only Shopify-specific gotchas Claude would not already know, or move them into a reference file, to tighten conciseness.
Add explicit validation checkpoints for write/deploy operations (e.g. 'After `shopify app deploy`, verify the extension target and confirm installation on the test store before proceeding') to lift workflow clarity above the destructive/batch cap.
Move the full GraphQL, TSX, and Liquid examples into their respective reference files and keep only a single representative snippet inline to improve content splitting and progressive disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The inline code blocks are tight, but the Best Practices section lists generic advice Claude already knows ('Use GraphQL over REST', 'Request only fields you need', 'Store API credentials in environment variables') and the trailing 'When to Use... applicable to execute the workflow' line is filler, matching anchor 3 (mostly efficient but could be tightened). | 3 / 5 |
Actionability | Provides concrete, mostly copy-paste-ready guidance — CLI commands, real GraphQL queries with variables, a TSX checkout extension, a Liquid snippet, and TOML config covering common cases — but has minor gaps such as the TSX example omitting the React `useState`/`useEffect` import, keeping it at anchor 4 rather than 5. | 4 / 5 |
Workflow Clarity | The ROUTING IF/THEN tree and sequenced CLI commands give a clear sequence, but there are no explicit validate-then-proceed checkpoints, and write/deploy/bulk operations (metafieldsSet, app deploy, bulk operations) lack feedback loops, so per the destructive/batch cap workflow clarity stays at 3. | 3 / 5 |
Progressive Disclosure | The overview signals one-level-deep references (app-development.md, extensions.md, themes.md), two scripts, and doc links with clear descriptions, but substantial detailed code (full GraphQL queries, TSX extension, Liquid snippet, webhook TOML) is inlined in SKILL.md rather than split into the reference files, matching anchor 4 ('key examples inline, bulk in separate file'). | 4 / 5 |
Total | 14 / 20 Passed |